"...As noted on the RC thread, it's really a shame that they didn't mix up the samples so they all had the same salinity. I can't fathom why they didn't..." / "...They did not mix to 35ppt SG..." You are mixing terminology. Mixing to a SG and testing was not the point of the test. The goal was to measure the amount of elements, believed important to us aquarists, in a given, quantified amount of mix.
"For example, 35 ppt value simply means 35 parts per thousand. If you broke down the water into all it's parts, out of a thousand of those parts, 35 would be salt (sodium chloride). Another way to think of it is simply 3.5%.
The 1.026 sg value is a little trickier. The sg stands for Specific Gravity. Specific gravity is not so much a measure in units, as it is a measure of density as compared against pure water. Pure water has a Specific Gravity (density) of 1.0. SG is a quick way to guestimate the salt content of your water. It is not always the most accurate way, however.
In practice, either value is intended to give you a measure of the concentration of the salt content of the water.
One common mistake made early on, is not taking into consideration the affect temperature has on salinity. Since temperature will make the water expand/contract, as a result, this will affect the density of the saltwater. There is a decent chart that can be found here, which will show the changes in salinity due to temperature." (
The Reef Tank - Powered by vBulletin ) If you consider the goal, which is to compare how much of each element is found in a given, standard sample, than you can conclude that mixing to a given SG is not productive as mixing to an unquantified sg will skew the amount of elements found in any given sample. The quote, however, does raise a *very* important point regarding temperature, which I do not see stated anywhere in the PDF. I do not know that temp would affect the element measurements, but it would have been nice to notate.
HTH, Enjoy!